This project is a Customer Churn Prediction System developed for a bank. It uses machine learning to predict whether a customer is likely to leave the bank (churn) or stay based on their historical data. The project leverages several libraries, including NumPy, TensorFlow, scikit-learn, Keras, and Matplotlib.
Customer churn is a critical issue for many businesses, including banks. By predicting which customers are at risk of leaving, banks can take proactive measures to retain them. This system uses machine learning algorithms to analyze customer data and predict churn.
- Data preprocessing and cleaning
- Feature engineering and selection
- Model training and evaluation
- Visualization of results
- Predicting customer churn
The dataset used for this project contains information about bank customers, including their demographic details, account information, and transaction history. The data is preprocessed and split into training and testing sets for model evaluation.
Clone the repository: (https://github.com/salimshakeel/Customer_churn-prediction_system.git)
The model's performance is visualized using Matplotlib. Various plots are generated to show the distribution of features, model accuracy, and other relevant metrics. Example results include: Accuracy: The overall accuracy of the model on the test dataset. Confusion Matrix: A visual representation of the model's performance. ROC Curve: The Receiver Operating Characteristic curve, showing the trade-off between true positive rate and false positive rate. Feature Importance: Visualization of the most important features contributing to the prediction.